{"id":"846d1473-fac5-492d-847c-25abbbb93d89","arxiv_id":"2507.18582","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Chiral superconducting trial wavefunctions (Pfaffian and K2a) have lower variational energy than an optimized quarter Fermi liquid in a rhombohedral graphene model near the flat-band regime.","lead":"The authors use variational Monte Carlo to compare trial chiral superconducting states against a spin-valley polarized Fermi liquid in a model of rhombohedral graphene. They find the superconducting trial states win in energy in a density window near a flat band bottom, suggesting repulsive Coulomb interactions alone can drive superconductivity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The phase diagram omits the Wigner crystal from the energy competition, and the ~1 meV condensation energy has no error bars; the claim that chiral SC is the relevant ground state above the WC density remains conditional.","rationale":"The paper makes a genuine numerical advance: the QFL Slater-Jastrow calculation is benchmarked against Ref. 38 within 0.2%, the new Pfaffian ansatz is shown to beat its Laughlin-type predecessor, and the K-matrix constraints ensure antisymmetric wavefunctions with finite kinetic energy per particle. The SC-vs-QFL comparison is internally coherent and honestly described. The load-bearing gap is that the advertised result is about the physical phase in the density window between the Wigner crystal and the QFL, but only the QFL boundary is computed. In a purely repulsive 2D system at these densities the Wigner crystal is the natural low-energy competitor, and the paper itself identifies its energetics as crucial. The ~1 meV condensation energy is small relative to typical variational bias; without error bars or finite-size extrapolation, one cannot exclude a reversal from either an improved QFL or the WC. The omitted long-wavelength density fluctuations and Berry curvature are acknowledged by the authors and argued to bias against SC, so they are less threatening to the direction of the conclusion; the WC energy is the unaddressed competitor that could go either way. This does not undermine the internal variational result that SC trials can beat the QFL trial, so REJECT would be too strong; it does mean ACCEPT is not warranted. The reader's CONDITIONAL verdict is appropriate.","tokens_in":19120,"tokens_out":13081,"duration_ms":147648,"concrete_test":"Perform VMC (or fixed-node DMC) for a triangular-lattice Wigner crystal with localized orbitals, the kinetic operator c2 k^2 + c4 k^4, and Ewald Coulomb interaction, at n_e = 0.2, 0.4, and 0.6×10^12 cm^-2 for the three parameter sets in Fig. 1. Compare the resulting energy per electron with the optimized Pfaffian and K2a energies. If the WC energy is below the best chiral SC energy by more than ~1 meV at any tested density, the abstract's claim that chiral SC wins above the WC density is not supported; if WC is consistently above, that condition is satisfied.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's only quantified variational competition is between chiral SC trial states and a Slater-Jastrow QFL. The Wigner crystal is never computed; Section VI says 'To complete the phase diagram, including the energetics of the Wigner crystal also will be crucial,' and Fig. 1's WC boundary is taken from experiment rather than from the model. Since the abstract and conclusion assert SC is favored 'above the density of Wigner crystal phase,' the central claim requires the model WC energy to lie above the SC energy in roughly 0.2-0.7×10^12 cm^-2. The claimed margin is only ~1 meV per electron (Fig. 4), while no statistical or finite-size error bars are given. Appendix A reports that the QFL trial Jastrow differs from the well-optimized Gaskell form by less than 10% of the correlation energy, so the QFL variational bias is not fully quantified. An improved QFL or a WC competitor could plausibly reverse the ordering even if the SC-vs-QFL comparison is internally consistent.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript uses variational Monte Carlo (VMC) to compare the energies of topological chiral superconducting trial states—a single-species Pfaffian state and two-species K2a/K2b Laughlin-type states—against an optimized Slater-Jastrow quarter Fermi liquid (QFL), for the dispersion E_k = c2 k^2 + c4 k^4 motivated by rhombohedral graphene. The authors introduce a modified Pfaffian ansatz (Eq. 7) and a two-parameter Slater-Jastrow QFL wavefunction (Eq. 12), optimize the variational parameters, and construct phase diagrams in density versus -c2 and in density versus magnetic field. They conclude that the Pfaffian and K2a chiral superconducting states can be energetically favored over the QFL at densities relevant to experiment (0.2–0.7 x 10^12 cm^-2), with a condensation energy of about 1 meV per electron, and they propose this as a mechanism for superconductivity from pure repulsive Coulomb interactions without a Fermi-surface pairing instability.","tokens_in":19377,"tokens_out":5283,"duration_ms":53890,"significance":"If the result is correct, it provides numerical support for a qualitatively new route to superconductivity in flat-band two-dimensional systems, driven by Coulomb repulsion through flux attachment rather than through a weak-pairing BCS instability, and it connects directly to the chiral superconducting phase observed in rhombohedral multilayer graphene. The paper's internal checks are a real strength: the QFL energies reproduce the Tanatar-Ceperley quadratic-dispersion result within 0.2% of the Coulomb energy, the alternative kinetic-energy estimators agree within 0.5%, the Hartree-Fock limit is recovered, and the new Pfaffian ansatz is shown to beat the previously used Laughlin-type ansatz. The important limitations—missing Wigner crystal energetics, missing error bars, and incompletely optimized chiral wavefunctions—are acknowledged by the authors, but they are precisely the quantities needed to make the phase-diagram claim quantitative rather than conditional.","major_comments":[{"comment":"The central claim in the abstract and conclusion that chiral superconductivity wins 'above the density of Wigner crystal phase' is not established by the calculations, because the Wigner crystal is never computed from the same Hamiltonian. Section VI states that 'including the energetics of the Wigner crystal also will be crucial,' and the Wigner crystal boundary in Fig. 1 is taken from experiment rather than from a variational energy of the model. Given that the reported SC-QFL margins are only about 1 meV, a Wigner crystal trial state could interpose or shift the phase boundary without contradicting any calculation in the paper. The phase diagram should either include a Wigner crystal energy calculation or the claims should be restricted to a comparison between the chiral SC states and the QFL.","section":"Section VI and Fig. 1"},{"comment":"No statistical or finite-size error bars are reported for the central energy differences shown in Fig. 4. The text describes substantial Monte Carlo samples (for example, 5 x 10^5 samples for the Pfaffian kinetic energy with N=70 and 3 x 10^6 samples for the two-species potential energy with N=200), but the condensation energy of approximately 1 meV is presented without an uncertainty estimate and without a thermodynamic-limit extrapolation for the chiral droplet states. Since the phase boundaries in Fig. 1 and the magnetic-field diagram in Fig. 5 are separated by energy differences of this size, the statistical significance of the ordering cannot be assessed from the data as presented.","section":"Fig. 4 and Section V"},{"comment":"The authors explicitly state that the optimized Pfaffian, K2a, and K2b wavefunctions 'do not incorporate the density fluctuations of the compressible superfluid mode' and that Berry curvature is omitted. These omissions are expected to lower the energies of the chiral states, but because the trial states are variational, the comparison to the QFL is not a complete energy competition. The likely improvement of the chiral states relative to the QFL is plausible but unquantified, and it could be comparable to the 1 meV differences reported in Fig. 4. A quantitative estimate of the missing chiral-state correlations, or a clear statement that the phase boundaries may shift when they are included, is needed.","section":"Section I and Eq. (7)"},{"comment":"The QFL side of the comparison also has unquantified variational bias. Appendix A reports that the trial Jastrow factor differs from the Gaskell form by less than 10% of the correlation energy; near the transition density the correlation energy is several meV, so this residual can be a substantial fraction of the claimed 1 meV SC condensation energy. The Gaskell comparison is made only for the quadratic-dispersion parameter set c2 = 91 meV nm^2, whereas the phase boundary in Fig. 4 uses c2 = -4 pi c4 n_e. The conclusion that an 'ideal' QFL would still lie above the chiral states therefore needs a quantitative propagation of the 10% residual into the phase diagram, not just the statement that the residual is less than 10%.","section":"Appendix A and Fig. 7"}],"minor_comments":[{"comment":"There is a typo in the displayed K matrix: the lower-right entry reads 'm.' and should be 'm'.","section":"Eq. (10)"},{"comment":"The parameters labeled 'xi_1' and 'xi_2' should be typeset as xi_1 and xi_2, and the sampling ranges written as '0.2 intervals' do not specify whether the grid is inclusive of the upper bound.","section":"Section II.A"},{"comment":"In the first sentence, 'naley' should be 'namely'.","section":"Appendix B.3"},{"comment":"In the paragraph discussing future directions, 'supercondcutor' should be 'superconductor'.","section":"Section VI"},{"comment":"The sentence beginning 'The indeed shows that as a positive quadratic term...' has an incomplete subject and should be rephrased.","section":"Section V"},{"comment":"The allowed occupation numbers n_k < 2 for two-species states and n_k < 1 for single-species states need a brief justification in terms of species degeneracy and spin/valley labels; as written, the reader may misread these bounds as a violation of Pauli exclusion.","section":"Appendix C"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [colleague],\n\nThe thing to know: this is a real VMC calculation, not a back-of-envelope argument, and the internal benchmarks are clean. The thing to be careful about: the central claim outruns the computation by a clear margin, so the right verdict is conditional, not accept.\n\nWhat is genuinely new: the optimized Slater-Jastrow treatment of the quarter Fermi liquid with quartic dispersion, the modified Pfaffian ansatz with algebraic long-range correlations, and the first energy comparison of those chiral superconducting trial states against an optimized QFL. The QFL benchmark against Tanatar-Ceperley is within 0.2% of Coulomb energy, kinetic sampling checks agree within 0.5%, and the Hartree-Fock limit is recovered. That is reproducible, checkable work, and the paper says plainly what it did and did not compute.\n\nWhere it gets soft: the SC condensation energy is about 1 meV per electron, and there are no quoted error bars or finite-size analysis on that central number. The Wigner crystal is never included in the energy competition; the phase diagram's WC boundary is imported from experiment, while the abstract and conclusion say the SC states win 'above the density of Wigner crystal phase.' Those two statements are not equivalent. The paper itself admits in Section I that the chiral trial states do not include the compressible superfluid mode's density fluctuations, and that Berry curvature is omitted; Section VI says WC energetics 'will be crucial.' Given the margin is ~1 meV, any one of those missing pieces could plausibly flip the ordering. Appendix A's claim that the QFL Jastrow is within 10% of the Gaskell correlation energy is reassuring, but it does not close the gap: 10% of the correlation energy can still be comparable to 1 meV in the region of interest.\n\nThe citation pattern is understandable—the trial states come from the authors' own Ref. 17—but the paper does not lean on that reference for the energetic comparison; the QFL side is benchmarked externally. That is not a circularity problem.\n\nWho is this for? People working on rhombohedral graphene superconductivity and on trial-wavefunction methods for strongly correlated flat bands. It deserves a serious referee. I'd send it out with a request for: WC energies or at least an honest statement that they are missing, statistical and finite-size error bars on the key energy differences, and a softened abstract if those are not forthcoming. As is, the SC-vs-QFL comparison is credible; the SC-vs-world conclusion is not yet.","headline":"A credible, well-benchmarked VMC comparison that overreaches from 'beats the QFL' to 'relevant ground state above the Wigner crystal.'","tokens_in":19893,"tokens_out":2981,"would_cite":true,"duration_ms":32000,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Pure Coulomb repulsion can stabilize topological chiral superconducting states without a Fermi-surface pairing instability.","keywords":["variational Monte Carlo","chiral superconductivity","topological order","Pfaffian wavefunction","K-matrix Laughlin states","quarter Fermi liquid","rhombohedral graphene","Coulomb interaction"],"falsifier":"Compute the Wigner crystal energy for the same parameters and densities; if it lies more than roughly 1 meV per electron below the optimized Pfaffian and K2a energies, the claimed ordering fails. Alternatively, run the same variational comparison with wavefunctions that include the compressible superfluid density fluctuations and Berry curvature; a shift of more than the roughly 1 meV condensation energy in either state's favor would overturn the conclusion.","tokens_in":18870,"feed_emoji":"⚡","tokens_out":9727,"duration_ms":89716,"temperature":0.7,"pith_summary":"Using variational Monte Carlo, this paper compares the ground-state energies of three topological chiral superconducting trial states (the single-species Pfaffian and the two-species K2a and K2b Laughlin-type states) against a fully optimized Slater-Jastrow description of the spin-valley polarized quarter Fermi liquid. For the dispersion $E_k = c_2 k^2 + c_4 k^4$, tuned to rhombohedral graphene parameters, it finds that the Pfaffian and K2a states have lower energy per electron than the Fermi liquid at densities up to about $0.5\\times10^{12}\\,\\mathrm{cm}^{-2}$, with a condensation energy near 5% of the Coulomb energy scale, roughly 1 meV per electron. The preference is strongest when $c_2$ is between zero and a negative value, i.e. just before a hole pocket forms at $k=0$. The paper reads this as evidence that superconductivity can emerge from purely repulsive Coulomb interactions in a nearly flat band, without a Fermi-surface pairing instability.","feed_headline":"No pairing needed: repulsion favors chiral superconductors","feed_subtitle":"Monte Carlo puts the Pfaffian and K2a states below the polarized Fermi liquid near ~0.5x10^12 cm^-2.","key_machinery":"The carrying object is the family of generalized Laughlin-type chiral superconducting wavefunctions built from holomorphic and antiholomorphic factors, with a K-matrix $K = K^+ - K^-$ whose diagonal entries are odd integers and whose null vector fixes the species density ratios through $\\sum_J K_{IJ} f_J = 0$, cancelling the macroscopic angular momentum that would otherwise make the kinetic energy diverge. For the one-species case the paper uses an improved Pfaffian ansatz $\\psi_{\\mathrm{Pf}} = \\mathrm{Pf}\\left(\\frac{1}{z_i-z_j}\\right)$ times a rotationally symmetric two-body product with four variational parameters, replacing the Gaussian decay of the Laughlin-type factor so that algebraic long-range density correlations are allowed. The comparison is made through the energy formula $E_{\\mathrm{tot}} = (e^2\\sqrt{n_e}/\\epsilon) V + c_2\\langle k^2\\rangle + c_4\\langle k^4\\rangle$, where $V$ is obtained from Monte Carlo pair-distribution functions and the kinetic averages from numerical derivatives of the wavefunction.","core_discovery":"The central claim is that strong repulsive Coulomb interactions alone can stabilize topological chiral superconducting phases over the spin-valley polarized quarter Fermi liquid in the density window relevant to the rhombohedral graphene experiments. After optimizing the variational parameters in the Pfaffian and two-species K-matrix wavefunctions, and in the Slater-Jastrow Fermi liquid, the Pfaffian and K2a states win with a condensation energy of about 1 meV per electron. The win survives up to densities around $0.5\\times10^{12}\\,\\mathrm{cm}^{-2}$, and it is largest for $c_2$ slightly negative, corresponding to a Fermi sea on the verge of developing a hole pocket at $k=0$. The paper argues these states are not BCS superconductors: they are driven by flux attachment and have a gapless superfluid density mode, short coherence length, broken time-reversal symmetry, and nontrivial topological order.","pith_inferences":["A testable extension is to track superconductivity against displacement field: this mechanism predicts the strongest chiral superconducting order immediately below the Lifshitz transition where $c_2 = -4\\pi n_e c_4$, and a rapid suppression once $c_2$ is clearly positive.","The paper did not include Berry curvature; since it expects that effect to lower the chiral states more than the Fermi liquid, including it would likely widen the superconducting region, with the K2b state benefiting most.","The trial chiral states omit density fluctuations of the compressible superfluid mode; a family that includes those fluctuations could lower the superconducting energies further, and comparing its effect state-by-state is a direct way to test whether the variational ordering is stable."],"forward_implications":["If the variational ordering is correct, the observed chiral superconductivity in rhombohedral multilayers should be understood as a strongly correlated topological superconductor rather than a BCS condensate of a Fermi surface.","A spin-unpolarized two-species state (K2a, in the same phase as a spin-triplet p+ip superconductor) remains competitive with the spin-polarized Pfaffian, so in-plane magnetic field robustness does not by itself rule out spin-unpolarized superconductivity.","The superconducting window is tied to a nearly flat band bottom: the energy advantage is largest when $c_2$ is slightly negative, just before a hole pocket forms, and shrinks as $c_2$ becomes positive.","The same pure-repulsion mechanism should be sought in other two-dimensional systems with an almost flat band bottom and strong Coulomb interactions, not only rhombohedral graphene."],"supporting_citations":[{"why":"Proposes the K-matrix Laughlin-type chiral superconducting wavefunctions and their zero-eigenvalue kinetic-energy constraint that this paper optimizes.","marker":"[17]"},{"why":"Supplies the experimental four-layer rhombohedral graphene parameters and density and magnetic-field windows that set the comparison regime.","marker":"[12]"},{"why":"Provides the benchmark quadratic-dispersion Fermi liquid energy that the paper extends to quartic dispersion and validates against.","marker":"[38]"},{"why":"Establishes the VMC and Ewald-sum methods for the correlated two-dimensional electron gas used for the Fermi liquid calculation.","marker":"[37]"},{"why":"Establishes the equivalence between the Pfaffian state and a spinless p+ip topological superconductor used to classify the one-species state.","marker":"[36]"},{"why":"Supplies the original Laughlin wavefunction and pair-distribution energy evaluation the two-species potential-energy calculation generalizes.","marker":"[48]"},{"why":"Provides the disk-geometry Monte Carlo evaluation of trial wavefunctions whose single-species potential energy is used as a sanity check.","marker":"[49]"}],"fun_headline_variants":["Pure repulsion yields chiral superconductor, no pairing","Coulomb repulsion alone stabilizes topological chiral superconductor","VMC finds repulsion-driven chiral superconductor beats Fermi liquid","Repulsion only: chiral superconductor without phonons","Flat band bottom: repulsion favors chiral superconductor"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument stands on the assumption, flagged in Section I, that the trial wavefunction family—the K-matrix Laughlin-type chiral states and the Pfaffian, plus the Slater-Jastrow Fermi liquid—contains the true ground-state correlations of the repulsive Coulomb Hamiltonian, so that the winner of the variational competition is the real phase. The paper itself notes in Section I that the superconducting trial states do not incorporate density fluctuations of the compressible superfluid mode, omit Berry curvature, and do not compute the Wigner crystal.","fun_headline_variants_meta":{"raw":{"variants":["Pure repulsion yields chiral superconductor, no pairing","Coulomb repulsion alone stabilizes topological chiral superconductor","VMC finds repulsion-driven chiral superconductor beats Fermi liquid","Repulsion only: chiral superconductor without phonons","Flat band bottom: repulsion favors chiral superconductor"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001209,"raw_usage":{"total_tokens":4965,"prompt_tokens":919,"completion_tokens":4046,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":535,"completion_tokens_details":{"reasoning_tokens":3962}},"tokens_in":535,"tokens_out":4046,"duration_ms":29619,"temperature":1.0,"reasoning_tokens":3962,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T18:10:20.108434+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the Wigner crystal energy for the same parameters and densities; if it lies more than roughly 1 meV per electron below the optimized Pfaffian and K2a energies, the claimed ordering fails. Alternatively, run the same variational comparison with wavefunctions that include the compressible superfluid density fluctuations and Berry curvature; a shift of more than the roughly 1 meV condensation energy in either state's favor would overturn the conclusion.","supporting_citations":[{"cited_title":"Wiegmann, Topological Superconductivity, Progress of Theoretical Physics Supplement107, 243 (1992)","cited_arxiv_id":null,"evidence_quote":"Provides the benchmark quadratic-dispersion Fermi liquid energy that the paper extends to quartic dispersion and validates against."},{"cited_title":"Wen and A","cited_arxiv_id":null,"evidence_quote":"Establishes the VMC and Ewald-sum methods for the correlated two-dimensional electron gas used for the Fermi liquid calculation."},{"cited_title":"Lee, Anyon superconductivity and the fractional quantum hall effect, Physica B: Condensed Matter169, 37 (1991)","cited_arxiv_id":null,"evidence_quote":"Establishes the equivalence between the Pfaffian state and a spinless p+ip topological superconductor used to classify the one-species state."},{"cited_title":"Wen and A","cited_arxiv_id":null,"evidence_quote":"Supplies the original Laughlin wavefunction and pair-distribution energy evaluation the two-species potential-energy calculation generalizes."}],"review_version":2}